PROGRESS REPORT Intravenous Hyperalimentation in Pediatrics John D. Lloyd-Still, MD, Harry Shwachman, MD and Robert M. Filler, MD

From the Departments of Medicine and Surgery, The Children's Hospital Medical Center, Boston, Massachusetts and Departments of Pediatrics and Surgery, Harvard Medical School. Supported in part by grants from the National Cystic Fibrosis Research Foundation and McGaw Laboratories, Glendale, California. Address for reprint requests: Harry Shwachman, The Children's Hospital Medical Center, 300 Longwood Avenue, Boston,Mass.

and their relation to total caloric intake. Efforts to provide amino acids for protein synthesis were pioneered in this country by Warren Cox, who provided casein hydrolysates in the form of Amigen ~ for experimental trial. Shohl et al (2) indicated in the early 1940's that positive nitrogen balance could occur in children provided with this form of nitrogen. It was clear that these experiments could not be continued for long because of untoward reactions, especially febrile reactions, and difficulties in providing adequate calories. It remained for Dudrick et al (3) to show that adequate calories could be provided by administering concentrated solutions of glucose into a central vein (an area of high blood flow). This surgical team working at the University of Pennsylvania provided practical guidelines in 1968 which are now being used throughout the country. Their experiments demonstrated, first in dogs and then in man, that growth, development, and positive nitrogen balance could be achieved by delivering nutrients intravenously. Six beagle puppies fed entirely intravenously for from 72 to 256 days were compared with an orally fed litter mate. The 6 puppies on intravenous feeding outstripped the orally fed controls in weight gain and matched them in skeletal growth, development and activity. Moreover, no nutritional abnormalities were observed in the parenteratly fed group, despite the length of the study. Dudrick et al also reported on clinical studies in 30 patients with chronic gastrointestinal disease. Included in their original paper was the report of an infant with multiple small bowel atresias, w h o u n d e r w e n t n u m e r o u s operative procedures. She was fed entirely by vein for 44 days and had normal growth and development, as indicated by increase in head

Digestive Diseases, Vol. 17, No. 11 (November 1972)

1043

Numerous attempts to provide intravenous nutrients to sustain life have been made in the past. Ideally, the ingredients would be fat, protein, carbohydrate, water, minerals and vitamins. Undoubtedly, whatever success we can claim today is a result of our improved technology and surgical skill, and the availability of commercial preparations and "setups" including pumps, filters, tubing, etc. Parenteral hyperalimentation can be defined as the intravenous administration of fluid, calories, nitrogen and other nutrients sufficient to achieve tissue synthesis and metabolism in patients of all ages with normal or excessive needs. HISTORICAL

ASPECTS

Early explorations in the use of fat emulsions were carried out in the mid-1930's by Emmett Holt and later by Stare, Geyer and associates, working at the Harvard School of Public Health (1). Although a great deal of scientific effort has gone into the provision of an adequate fat emulsion for intravenous use, none has fulfilled the requirements of the Federal Drug Administration, so no preparation is available for the clinician in this country. Wretlind of Sweden has not only contributed to the preparation of a clinically available fat emulsion but also studied the amino acid requirements

LLOYD-STILL ET AL

circumference, length and body weight. This same infant's case history was amplified in a subsequent publication (4). Filler and associates (5) in 1969 at T h e Children's Hospital Medical Center in Boston described their initial experience with longterm parenteral hyperalimentation in 14 critically ill infants under 2 months of age in whom all oral intake was completely withheld for as long as 60 days. T h e use of the so-called "lifeline" (as this procedure came to be known) was extended by Shwachman to the management of infants with protracted diarrhea and severe malnutrition. A report of this experience was p r e s e n t e d at the E u r o p e a n C o n f e r e n c e of Pediatric Gastroenterology in 1969 (6). This form of therapy was life saving in these infants, the majority of whom had been referred from outside hospitals because they failed to respond to conventional therapy.

tervals. Second, it was difficult to deliver intravenous fluids in sufficient volume and conc e n t r a t i o n w i t h o u t e i t h e r t h r o m b o s i n g the peripheral veins or overloading the circulation. The unique contribution of the Philadelphia workers was to devise a technic whereby adequate caloric intake could be maintained by administering hypertonic glucose solutions through the superior vena cava. To ensure proper use of large quantities of glucose, the infusate was given at a slow rate regulated by a constant infusion pump. T h e risk of sepsis was minimized by tunneling the catheter from its venous entry site to a skin exit site 4 to 5 cm distant. In addition, a bacterial filter was added to the system and a program of meticulous catheter care was instituted. With this technic it was possible to obtain positive nitrogen balance even after extensive surgical procedures.

RATIONALE

I N D I C A T I O N S FOR I N T R A V E N O U S HYPERALIMENTATION

Knowledge of fluid and electrolyte balance, together with coincident advances in anesthesia and surgical technics, now brings about the survival of many infants and children under conditions that would formerly have been fatal. However, despite initial recovery from the surgical procedure, poor caloric intake and negative nitrogen balance may later result in nutritional failure. These complications are especially prone to occur after operations on the gastrointestinal tract of the neonate or younger infant. Moreover, in developed countries, nutritional failure and secondary infection have become the most common causes of death in infants with persistent diarrhea. Under circumstances where enteral feeding is precluded, all the nutritional requirements of the individual must be administered parenterally. In the past, two major complications of parenteral hyperalimentation have limited its application for more than short intervals. First, there was considerable risk of sepsis from the use of an indwelling catheter for prolonged in-

Intravenous hyperatimentation is indicated when prolonged adequate nutrition by the oral route is impossible or hazardous. It may be started prophylactically to avoid malnutrition associated with those conditions in which clinical experience has shown that the gastrointes-tinal tract will not be functional for periods longer than 1 week. If the intestinal tract can tolerate special oral or tube feedings, hyperalimentation should not be attempted. Intravenous hyperalimentation should not be started, or should be discontinued, if the patient has a gastrointestinal tract which will never function adequately despite all forms of therapy. This method of feeding should not be used to unnecessarily prolong suffering or to continue a meaningless existence. Some of the specific indications for the use of parenteral hyperalimentation in pediatrics are shown in Table 1. About two-thirds of the total series of over I10 cases at T h e Children's Hospital Medical Center in Boston were children who also required surgery. This is in con-

1044

Digestive Diseases, Vol. 17, No. 11 (November 1972)

INTRAVENOUS HYPERALIMENTATION Table 1. Indications for Parenteral Hyperalimentation in the Pediatric Age Group Surgical Omphalocele Esophageal abnormalities Bowel atresias Volvulus Meconium ileus Other congenital malformations Perforations Fistulas Short-bowel syndrome Burns Multiple trauma Medical Protracted diarrhea of infancy Nutritional failure Ulcerative colitis Crohn's disease Tumor enteritis

Malignancies Investigational Coma Heart disease Prematurity Hepatic failure Renal failure Psychiatric disorders Metabolic disorders

trast to adults where almost 50% of the patients treated with parenteral hyperalimentation had medical causes for their nutritional failure. Chronic protracted diarrhea of infancy is the most common medical indication for parenteral nutrition in the pediatric age group. T h e majority of our patients were infants under 4 months of age. Pathogenic Esckerichia coli were isolated in only 3 of 16 patients although a pathogenic organism was isolated in 44% of the total group (7). Prior to the advent of parenteral hyperalimentation, most of these infants would have succumbed from inanition, nutritional failure, and terminal sepsis. T h e total length of hospitalization for those who survived in the past was usually several months. Autopsies on children who died from this condition usually showed extensive enterocolitis (8). In most of our patients, per oral biopsy of the small intestine has shown marked atrophy of the mucosa accompanied by severe secondary disaccharidase deficiency. However, in others the histology and disaccharidases have been more nearly normal, and some form of enterotoxin may have been responsible for the protracted diarrhea and nutritional failure. T h e

Table 2. Details of Constituents of Hyperalimentation ("Lifeline") Solution Ingredients

Strength

Amount(ml)

Manufacturer

1. 2. 3. 4. 5. 6.

Hyprotigen 5% D5% Dextrose 50% Potassium phosphate 2 mEq/ml Aqua mephyton* 0.2mg/ml Folic acid* 0.5 mg/ml MgSo 4 • 7H20 10%

910 500 3.0 1.5 1.5 12.0

McGaw Labs, Glendale, Calif 91201 McGawLabs, Glendale, Calif 91201 McGaw Labs, Glendale, Calif 91201 Merck, Sharp & Dohme, West Point, Pa 19486 Lederle Labs, Pearl River, NY 10965 Lilly and Company, 307 E. McCarty St, Indianapolis, Ind 46206

7.

MV.I. (Multivitamin Solution)t Calciumgluconate 10% Vitamin B12 10 mCg/ml

7.5 75.5 1.0

US Vitamin and Pharmaceutical Corp. New York 10701 Upjohn Company, Kalamazoo, Mich 49001 Squibb and Sons, 909 Third Ave, New York 100.22

8. 9.

Procedure: Add nutrients 3 through 9 to 500 ml of 50% dextrose in water in the order listed. This 602 ml of solution is divided evenly into 3 evacuated containers and each filled to 500 ml with 5% Hyprotigen in 5% dextrose. The final solution is shaken and used within 24 hours of opening, Calories:O.77/ml. *Indicates dilution from manufacturer's concentration tAscorbic acid (C), 500 mg; vitamin A, 10,000 U.S.P. Units: vitamin D (ergocalciferol), t,O00 U.S.P. Units; t h i a m i n e H C l ( B t ) , 5 0 m g : r i b o f l a v i n (B2), 10 rag: niacinamide. 100 rag: pyridoxine HCI (B6), 15 rag; dexpanthenol, 25 rag: vitamin E (dl-~ tocopheryl acetate), 5 IU.

Digestive Diseases, Vol. 17, No. 11 (November 1972)

1045

LLOYD-STILL ET AL Table 3. Final Composition of Hyperalimentation Solution Each 1000 ml contains: Protein hydrolysate (108 calories) 30.0 g Dextrose (hydrous) (668 calories) 196,6 g Potassium 16,0 mEq Sodium 15.0 mEq Calcium 27.0 mEq Phosphorous (HP04) 19.0 mEq Magnesium 7.68 mEq Chloride 10.8 mEq Folic acid 0.5 mg Multivitamins (M.V.I.) 5.0 ml Vitamin K 1 0.2 mg Vitamin B]2 6.6#g Each 1000 ml contains approximately 4.8 g nitrogen and 776 calories.

decision as to when to use hyperalimentation is a clinical one. The age of the patient, degree and duration of nutritional failure, persistence of diarrhea in spite of parenteral fluids, inability to correct hypoproteinemia, and development of deficiencies such as hypoprothrombinemia are taken into account in making a final decision. Chronic inflammatory bowel disease is being increasingly recognized in the pediatric age group. Some children with Crohn's disease and ulcerative colitis develop nutritional failure, especially when their disease is longstanding. Parenteral hyperalimentation may be indicated in these patients as a preoperative procedure, and striking improvement in the patient's wellbeing may ensue, especially in psychologic status (9). Extensive resections of involved bowel may result in the development of the shortbowel syndrome. With parenteral hyperalimentation, it is possible to put the gastrointestinal tract at rest, while positive nitrogen balance is achieved and nutritional deficiencies are remedied. In some patients, remissions have coincided with the course of hyperalimentation. Complications such as fistula formation and toxic megacolon, which are usually considered indications for surgery, have sometimes subsided on parenteral hyperalimentation. 1046

Patients with malignant disease can develop nutritional failure from a variety of causes; they account for another large group that may benefit from parenteral hyperalimentation. Therapy with cytotoxic drugs and irradiation is usually accompanied by anorexia and vomiting; a more specific entity, tumor enteritis, ensues when the bowel itself is directly damaged by therapy. COMPOSITION

OF I N F U S A T E

As stated previously, intravenous fat emulsions are not presently available for clinical use in the United States. Actual volumes of the various constituents used at the present time in the preparation of the infusate are given in Table 2. Modifications are made from time to time, with increased experience and knowledge of nutritional requirements. Mixing is done under a laminar flow filtered air hood, and sterility testing is essential. The final composition of the hyperalimentation solution is given in Table 3. The solution contains approximately 20% glucose and 3% protein in the form of protein hydrolysate (enzymatically hydro-

Table 4. Complications of Parenteral Hyperalimentation in the Pediatric Age Group Catheter Arrythmia Incorrect position (liver, heart) Slippage Thrombosis Atrial rupture Infective, local and systemic Bacterial Fungal Metabolic Osmotic diuresis Hyperglycemia Hypoglycemia Hypolipemia Hypocalcemia Hypercalcemia Hypomagnesemia Hypophosphatem ia Hypocupremia

Digestive Diseases, Voh 17, No. 11 (November 1972)

INTRAVENOUS HYPERALIMENTATION

®

-

. . . .

- ~

i

7

Fig. 1. A system for long-term intravenous alimentation, "lifeline." 1. Amino acid-glucose infusate. 2. Calibrated buret. 3. Constant infusion pump. 4. Disposable tubing with a compressible section which adapts to pump head. 5. Millipore filter. 6. T connector. 7. Silicone rubber intravenous catheter. [From Filler RM, Eraklis AJ (10)]

lysed casein). The caloric value is 0.77 calories/ ml. Usually, sodium, potassium, and chloride are added to the final solution at the bedside (10). Not only is the obligatory excretion of potassium in the urine usually increased, but also increased p o t a s s i u m r e q u i r e m e n t s are necessary to achieve weight gain and protein synthesis. Thus potassium is required in larger doses than with standard intravenous therapy. In general, 30 mEq of NaCI and 20 mEq KCI are added to each liter to raise the concentration of sodium to 43 mEq/liter, of potassium to 36 mEq/liter, and of chloride to 67 mEq/liter. In young infants, or when renal function is im-

paired, it is usual to start with half-strength hyperalimentation solution at a rate of about 60 to 70 ml/kg/24 hr. By this means excessive glycosuria, leading to osmotic overload, is avoided. Within 24 to 48 hours, the volume and strength of the infusate is increased until fullstrength solution is given at a rate of about 135 ml/kg/24 hr. This solution supplies 104 calories/kg/day. Calcium, phosphorus and magnesium are always added to the infusate for the infant or growing child. Daily requirements of all vitamins are included in the infusate. Plasma infusions (10 ml/kg) are given twice weekly in an

Digestive Diseases, Vol. 17, No. 11 (November 1972)

1047

LLOYD-STILL ET AL

attempt to provide essential fatty acids and trace metals. Blood transfusions are also given when indicated. Iron therapy is indicated if parenteral nutrition is continued for a prolonged time.

the bacterial filter. A calibrated buret is placed in line to accurately monitor the volume delivered. If intravenous injection of a solution distal to the bacterial filter is required, a " T connector" can be inserted in the line. The entire system for intravenous nutrition is shown in Figure 1.

TECHNIC

T h e central venous catheter is usually inserted through the external or internal .jugular vein in the operating room with the use of local or general anesthesia. A small *ransverse. incision is made over the sternomastoid muscle in the neck and the appropriate .jugular vein prepared for cannulation. Before passing the catheter into the vein, the catheter is tunneled subcutaneously from the vein entry point to the scalp. A skin exit site at a distance from the venous intubation facilitates proper cleansing of the skin and minimizes the possibility of accidental removal of the catheter. Silicone rubber catheters are softer and less reactive than polyvinyl or polyethylene tubes and are now used routinely. T h e catheter is inserted into the jugular vein and advanced to the superior vena cava at the level of the right atrium. The exact location of the tip should be confirmed radiographically by filling the catheter with radiopaque contrast material. Antibiotic ointment (Betadine ~) and a sterile dressing are applied to the skin exit site; every 3 days this dressing is removed meticulously, the skin cleaned with Betadine solution, and a sterile dressing and antibiotic ointment reapplied. T h e catheter is not changed unless it is accidentally occluded or dislodged or sepsis develops. Ordinarily the line is not used for blood sampling; this avoids the hazards of catheter thrombosis and infection. Before the infusion of the nutrient solution, a Millipore -~ bacterial filter is placed in the circuit to remove particulate matter a n d / o r microorganisms which might have contaminated the solution. T h e infusate is delivered by a variable-rate infusion pump to insure uniform delivery and to overcome resistance offered by

T h e following clinical measurements are considered essential in evaluating the patient's metabolic response--daily body weight, accurate urine volume, other body fluid losses. Laboratory measurements include qualitative urinary sugar, blood sugar concentrations, serum electrolytes and osmolarity. In the stable patient, the urine is tested for sugar at each voiding. Measurements are repeated at 3-day intervals. Estimation of total protein, albumin, prothrombin time, calcium, liver function tests and magnesium will be necessary at intervals in patients with persistent diarrhea. Weight gain will depend on the initial clinical condition of the patient. Weight gains may be expected to proceed at a normal rate in infants who are not malnourished at the time that intravenous fluids are introduced. When infection is present or when metabolic requirements are increased, the growth curve may remain stationary or rise more slowly. In children who are severely depleted it may take up to 2 weeks for a significant .weight gain to occur. Despite the variations in weight curves, positive nitrogen balance has been observed in all patients studied. On intravenous hyperalimentation which provides 0.74 g nitrogen/kg/day (equivalent to 4.4 g protein/kg/day) persistent positive nitrogen balance of 100 to 300 g / k g / d a y has been observed. Metabolic studies have shown that, even with the administration of a large quantity of glucose (27 g / k g / 2 4 hr), when infused at a constant rate, the blood sugar levels usually range between 44 and 108 mg% (11). Glycosuria varied between 0 and 4 + by the clinitest method.

1048

Digestive Diseases, Vol. 17, No. 11 (November 1972)

OBSERVATIONS

INTRAVENOUS

HYPERALIMENTATION

5.5-

"J --

~-

0~

5.0WT. 4.5IN KGMS 4.03.5"

NO.OF STOOLS

10

m m I ~ Fig 2. Patient JR. Protracted diarrhea of infancy treated by parenteral hyperalimentation, "lifeline," for 29 days. ~ = 3200-AD: Special Formula, Mead Johnson & Comparny.

PLASMA I.V. FLUIDS BREAST

•

.

.

.

.

.

.

.

.

.

.

•

I

m

• !t 3200 AD NUTRAM I GEN NEOMYCIN RO.

i

i

t

BIRTH

I 20

I 40

/ 80

v 80

I 100

1 120

AGE IN DAYS

However, quantitative sugar analysis showed t h a t less t h a n 1% of t h e total glucose infused was lost in the urine. U r i n a r y s u g a r levels are g e n e r a l l y h i g h e s t d u r i n g t h e first day or t w o of t r e a t m e n t . In the m a n a g e m e n t of these children, qualitative urine sugars consistently above 3 + suggest an o s m o t i c diuresis. A t e m p o r a r y d e c r e a s e i n t h e h o u r l y infusion r a t e o r the use of a m o r e dilute solution will correct this problem. W a t e r r e q u i r e m e n t is met by p r o v i d i n g 135 m l / k g / d a y even t h o u g h the infusion solution is h y p e r t o n i c (2600 m O s m / l i t e r ) . Special a t t e n tion should be given to fluid b a l a n c e w h e n r e n a l function is i m p a i r e d .

CASE REPORT JR weighed 3.6 kg at birth. The neonatal course was uneventful and he was breastfed. From Day 4 he had eight loose stools daily, but weight gain was satisfactory, reaching 4.4 kg at 4 weeks. At this time, watery diarrhea developed, with occasional blood streaking in the stools. This continued over the next 10 days, until fever, dehydration and weight loss led to admission to a local hospital. ExaminaD i g e s t i v e D i s e a s e s , Vol. 1 7 , N o . 1 1 ( N o v e m b e r

•

1972)

tion at that time showed a severely ill, lethargic infant with a temperature of 102° F and an estimated 15% dehydration. Stool culture grew pathogenic E coli 0127; he was treated with a 5-day course of oral Neomycin~; six subsequent stool cultures grew no pathogens. Urine and stools showed 3+ reducing sugars. Owing to continuing diarrhea and failure to control the electrolyte abnormalities, he was transferred to The Children's Hospital Medical Center in Boston. With conventional therapy of intravenous fluids and withholding oral feedings, he slowly improved. After 7 days he was given small amounts of 3200-AD,* a special formula of glucose, medium-chain triglycerides, and casein hydrolysate. Although there was an initial improvement, he suffered another severe relapse 7 days after admission. By this time, his weight had fallen below the third percentile and his nutritional status had progressively deteriorated. The total protein was 3.1 g/100 ml and the albumin was 1.59 g/100 ml. Intravenous hyperalimentation was therefore instituted. On this regimen, his diarrhea ceased; weight remained stationary for 1 week and then rapidly increased. After 2 weeks of nothing by mouth, 3200-AD was again slowly commenced and was well-tolerated. Intestinal biopsy was performed 1 week prior to removal of the lifeline. (See Figure 2.)

*Mead .Johnson & Company, Evansville, Illinois. Now marketed as Pregestimil ~. 1049

LLOYD-STILL ET AL Findings were compatible with chronic enteritis and secondary disaccharidase deficiency. (Lactase 0.6; sucrase 1.83; mahase 5.78; palatinase 0.61 /*moles disaccharide split/min/g wet wt. Normal controls are lactase 3.6 4- 1.2; sucrase 7.4 4- 3.2; mahase 24.4 4- 9.4; palatinase 1.9 :~ 0.8 umoles disaecharide split/min/g wet wt.) In view of these findings, he was maintained on the lactose-free formula Nutramigen ~. Subsequent progress and development have been normal.

disproved. Pyrogen reactions with present-day solutions are virtually nonexistent. Use of the "lifeline" for blood sampling is inadvisable (except when sepsis is suspected) for this is another potential source for the introduction of bacteria. Although the bacterial filter in the system affords some protection against contamination of the infusate, some danger of infection from this source always remains.

COMPLICATIONS Infection

Technical

Sepsis still remains the major complication from the use of parenteral h y p e r a l i m e n t a t i o n in pediatrics. Moreover, under circumstances where strict adherence to the standard technic of catheter care declines, the incidence of infective complications rises sharply. M o s t studies on the infective complications following the long-term use of indwelling catheters gives an incidence of between 25 to 45%. In some of the early series of debilitated infants treated by the "lifeline" there was a similar incidence, but subsequent experience has brought about improvement. Once an infection has been documented, it becomes imperative to remove the catheter and institute a p p r o p r i a t e antibiotic therapy. Some authors have found that the appearance of glycosuria in infants on the lifeline can be an indication of impending sepsis. In 16 infants with chronic protracted diarrhea, 5 (31%) developed sepsis while on the lifeline, but no fatality occurred (7). T h e overall incidence of sepsis in 110 patients treated at T h e Children's Hospital Medical Center in Boston was 16 out of 110 (15%). Yeasts were found in twothirds of our patients with sepsis, and G r a m positive and G r a m - n e g a t i v e bacteria in the remainder. Infective complications of the lifeline accounted for 1 immediate death and later contributed to the demise of 4 other children. T h i s danger is by no means eliminated by the new technic, and it is imperative to have justifiable indications before embarking on a regimen of parenteral nutrition. W h e n fever occurs in a patient on the lifeline, the presumptive diagnosis must be sepsis until

M a n y of the early technical problems associated with the catheter have now been either corrected or reduced in frequency. T h e use of polyvinyl catheters was complicated by a high incidence of sterile inflammation and thrombosis and in some series by venous perforation. Soft, nonreactive silicone rubber catheters have minimized these hazards. T r a n s i t o r y arrythmias may occur during insertion of the catheter, especially with excessive manipulation. T h e position of the catheter tip must be checked radiographically to be certain that the irritating h y p e r o s m o l a r solution is delivered into the superior vena cava and not into a vein with a lower blood flow such as the hepatic. T h e soft silicone catheter must be properly secured to avoid dislocation during prolonged therapy.

1050

Venous T h r o m b o s i s of the catheter can occur, especially in the critically ill patient with sepsis and inadequate circulation. W h e n thrombosis occurs in the superior vena cava, it is better tolerated than in most other vessels. Although evidence of thromboembolism has been noted at autopsy in children dying with the catheter in place, manifestations of this phenomena in vivo are an extreme rarity. Probably the most important factor in the prevention o f v e n o u s complications at present lies in the x-ray control of the catheter position. Metabolic T h e problems of hyperglycemia and osmotic diuresis have already been discussed in "ObDigestive Diseases, Vol. 17, No. 11 (November 1972)

INTRAVENOUS HYPERALIMENTATION

servations." Hypoglycemia remains a hazard if there is a sudden cessation of the high glucose infusion. The volume of the infusate should be gradually reduced coincident with the onset or increase of oral feedings several days prior to removal of the lifeline. With this regimen hypoglycemia has not been a problem. In infants, hepatomegaly may develop during the initial period of parenteral nutrition but usually subsides within a few weeks. Hypolipemia is a constant finding, as fats are absent from the infusate; however, essential fatty acids are supplied by biweekly plasma infusions (10 ml/kg). When calcium was excluded from the infusate, hypocalcemia and tetany have resulted. Rickets and bone fractures have also been reported. At the other extreme, hypercalcemia has been encountered in the younger infant who is unable to utilize the quantity of calcium added to the infusate. Hypomagnesemia, characterized by hypothermia and seizures, occurred in 1 of our patients with longstanding chronic inflammatory bowel disease. Hypocupremia may contribute to the persistent anemia that is refractory to iron therapy and is usually associated with neutropenia and osteoporosis. Hypophosphatemia is known to interfere with red cell metabolism. Excess amlnoaciduria (especially glycine) may be seen, but balance data usually show that tess than 2% of the essential amino acids are excreted in the urine. Renal tubular reabsorption is usually excellent and plasma amino acid levels remain within normal limits, except in the premature infant when immaturity of liver enzyme systems presumably interferes with amino acid utilization (12). Finally, the possible deleterious effects of other metabolic changes such as hyperammonemia remain to be fully elucidated (13). CONCLUSION

Parenteral hyperalimentation has now found widespread acceptance and use in the treatment of severe nutritional failure in centers throughout the world. After more than 4 years of experience, the incidence of major complications Digestive Diseases, Vol. 17, No. 11 (November 1972)

from the procedure have been more clearly defined. Specific nutritional deficiencies are still being encountered, and it will be noted that many of the constituents in today's solutions have been modified from those described in the original reports. When using this form of therapy, it is certainly advisable to follow the latest developments in the investigative field. When contemplating parenteral hyperalimentation, one must be aware of the underlying condition of the patient. The possibility of a correctable surgical condition responsible for the nutritional failure should always be considered. P a r e n t e r a l h y p e r a l i m e n t a t i o n has made a significant contribution to the prevention and management of nutritional failure in patients who otherwise would have succumbed. However, the procedure carries welt-defined risks which must be carefully considered before embarking on this form of therapy. REFERENCES

1. Geyer RP: Parenteral nutrition. Physiol Rev 40:150, 1960 2. Shohl AT, Butler AM, Blackfan KD, MacLachlan E: Nitrogen metabolism during oral and parenteral administration of the aminoacids of hydrolysed casein. J Pediat 15:469, 1939 3. Dudrick SJ, Wilmore DW, Vats HM, Rhoads JE: Long-term total parenteral nutrition with growth, development and positive nitrogen balance. Surgery 64 : 134, 1968 4. Wilmore DW, Dudrick SJ: Growth and development of an infant receiving all nutrients exclusively by vein. JAMA 203:140, 1968 5. Filler RM, Eraklis AJ, Rubin VG, Das JB: Long-term total parenteral nutrition in infants. N Engl J Med 281:589, 1969 6. Shwachman H, Filler RM, Khaw KT: A new method of treating malnourished infants with severe chronic diarrhea. Acta Paediatr Stand 59:446, 1970 7. Lloyd-Still JD, Shwachman H, Filler RM: Protracted diarrhea of infancy treated by intravenous hyperalimentation: I. Clinical studies on 16 infants (In press) 8. Hyman CJ, Reiter J, Rodnam J, Drash AL: Parenteral and oral alimentation in the treat1051

LLOYD-STILL ET AL

ment of the non-specific protracted diarrheal syndrome of infancy. J Pediatr 78 : 17,1971 9. Dudrick SJ, Long JM, Steiger E, Rhoads JE: Intravenous hyperalimentation. Med Clin North Am 54:577, 1970 10. Filler RM, Eraklis AJ: Care of the critically ill child: intravenous alimentation. Pediatrics 46:456, 1970 11. Das JB, Filler RM, Rubin VG, Eraklis AJ: In-

travenous dextrose-amino-acid feeding: the metabolic response in the surgical neonate. J Pediatr Surg 5 : 127, 1970 12. Ghadimi H, Abaci F, Kumar S, Rathi M: Biochemical aspects of intravenous hyperalimentation. Pediatrics 48:955, I971 13. Heird WC, Driscoll J M Jr, Schullinger JN, Grebin B, Winters RW: Intravenous alimentation in pediatric patients. J Pediatr 80:351, 1972

1052

Digestive Diseases, Vol. 17, No. 11 (November 1972)